Free FLAC transcription, no sign-up

FLAC to Text Converter Lossless In, Text Out

Upload a FLAC file from a field recorder, an archive or a studio session and read it as text with a timestamp on every segment. Large files are uploaded in parts, so there is no need to shrink them to MP3 first. 10 transcriptions are free, each up to 3 hours long; guests read a free preview and a free account keeps the full transcript.

Drop your file or browse

MP3, M4A, WAV, OGG, FLAC, MP4, MOV, MKV, WEBM and more

Large video? Only the sound is needed for a transcript. Extract the audio first, or drop the video here and we will offer it.

10 free transcriptions, no account needed. Each recording can be up to 3 hours long, 5 hours in total, and a free preview of the transcript opens right here in your browser, with the full transcript in a free account. We run the models on our own servers, so a long recording can take a few minutes. If you create a free account later, everything you transcribed here moves into it, along with whatever is left of your free hours.

Need more than the free allowance? See the plans

1

Upload the FLAC file

Drop it into the box above. 16-bit and 24-bit files, mono or stereo, at 44.1, 48 or 96 kHz are all read. A big file takes a while to upload on a slow line; the upload runs in parts.

2

The recording is transcribed

The audio is decoded on our servers and passed to a speech model. The language is detected from the recording. Processing takes about three to five minutes per hour of audio.

3

Work with the text

Read and copy the transcript in the browser. With a free account the full text is saved, and you can correct names and terms in the editor.

Who has speech in FLAC

FLAC is best known as a music format, but a lot of spoken audio is stored in it. Oral history projects, university archives and libraries digitise tapes to FLAC because it is lossless, open and takes about half the space of WAV. Researchers who record interviews on Zoom, Tascam or Sony field recorders often choose FLAC on the device for the same reason.

Podcast and audiobook producers keep their masters in FLAC and publish MP3 copies. Some call recording systems and conference platforms export FLAC, and radio stations archive programmes in it. In all of these cases the file is the best available version of the recording, and it makes sense to transcribe that version, not a compressed copy.

Does lossless audio give a better transcript?

A little, and less than people expect. A speech model listens to a band of frequencies that a decent MP3 also preserves, so a clean interview transcribes almost the same from FLAC and from a 128 kbit/s MP3. The 24-bit depth and 96 kHz sampling of a studio file carry detail above what speech recognition uses.

The gain shows on difficult material: a quiet speaker far from the microphone, a digitised cassette with hiss, a room with echo. Lossy compression smears exactly those weak signals, and FLAC keeps them. So if you hold the FLAC, upload the FLAC. Converting it to MP3 to save upload time is reasonable on a very slow connection, at a bitrate of 128 kbit/s or more, but it never improves the result. Microphone distance, background noise and people talking over each other still decide most of the accuracy.

File size and long recordings

An hour of stereo FLAC at CD quality is roughly 250 to 350 MB, and a 24-bit, 96 kHz session can pass a gigabyte. Files are sent in parts, so a long upload is not lost to a single hiccup in the connection. For speech, a mono file is enough and half the size; if your recorder wrote two identical channels, exporting mono before uploading saves time without changing the transcript.

Archive recordings are often long: a two-hour life-story interview, a full day of a conference split into sessions. Each file can be up to 3 hours long as a guest. Timestamps on every segment let you cite a passage with its position in the recording, which is what catalogues and research notes usually need.

Free allowance and confidential material

You can transcribe 10 recordings without an account, 5 hours in total, and read a free preview of each. A free account uses the same allowance and keeps the full transcripts with an editor and a history. Summaries, AI chat with the transcript, export to PDF and Word and subtitle files in SRT, VTT or ASS are part of the paid plans.

Interview archives and research recordings usually come with consent forms and ethics approvals that limit where the audio may go. Transcription runs on our own servers with our own speech models, and recordings are not passed to a third-party speech API.

Frequently asked questions

Do I have to convert FLAC to MP3 or WAV first?

No. FLAC is decoded directly. Converting to WAV changes nothing except the size, and converting to MP3 throws away detail that can help on quiet or noisy recordings.

Is there a file size limit for FLAC?

The limit that matters is the length: up to 3 hours per recording as a guest. Large files are uploaded in parts, so a file of several hundred megabytes goes through, given a stable connection.

Are 24-bit and 96 kHz files supported?

Yes. High-resolution FLAC is accepted and resampled for the speech model. The extra resolution does not make the transcript better than a 16-bit, 44.1 kHz file of the same recording.

Can I transcribe a song in FLAC to get the lyrics?

This tool is built for speech. Sung vocals over instruments transcribe poorly, with missing and wrong lines, so we do not recommend it for lyrics.

Will the transcript mark who is speaking?

Guests get the text with timestamps on each segment. Check the plan details on the pricing page for what is included once you have an account.

Transcribe your FLAC recording

Scroll to the top and drop the file in at full quality. 10 transcriptions of up to 3 hours each are free and start without an account.